most cases the site mean also was not different. For example, the sediments from all nine stations in 

 western Long Island Sound (sites 4, 5, and 6) were not different from controls. Correspondingly, none 

 of these three site means was different from controls. Similarly, all nine stations and all three site 

 means in Newark Bay/ Arthur Kill were significantly different from controls. However, there were 

 some deviations from these patterns. For example, all three stations sampled at site 13 were not signifi- 

 cantly different from the controls, but the site mean was significantly different. The same situation 

 occurred in sites 29, 33, 37. The variances among the five replicates tested for each station were high, 

 but the variances among the three stations sampled at these sites were small, resulting in a significant 

 difference from the controls. Conversely, all of the stations at sites 10, 20 and 39 were significantly 

 toxic, but due to high variability among stations, the site means were not different from the controls. 



Eight of the nine samples tested with the freshwater amphipod Diporeia spp. by the Great Lakes Envi- 

 ronmental Research Laboratory (Dr. Peter Landrum) were significantly more toxic than controls (Table 

 6). Mean percent survival in Florissant Soil controls ranged from 88.7 to 100 (n=4, 4 or 8 replicates 

 each). The sample from station 38-A was nontoxic in both amphipod tests. Among the eight samples 

 toxic to Diporeia spp., seven also were toxic to Ampelisca abdita. Sample 9-B caused zero survivors in 

 both tests. Avoidance of all but samples 38-A and 6-B was significant relative to controls. Avoidance 

 was greatest of samples 18-B and 9-B. 



The results of the amphipod toxicity tests performed during Phase 2 with 57 samples are summarized 

 in Table 7. All except 6 samples were tested by SAIC in Narragansett, R. I. Because of the suspected 

 hazardous condition of samples 7-A, 7-B, 7-C, 8-A, 8-B, and 10, they were tested separately by Aqua 

 Survey, Inc. in Flemington, N. J. Tests organisms used by both SAIC and Aqua Survey were obtained 

 from the same source. In the controls and 6 test samples, water pH ranged from 7.7 to 8.6, temperature 

 ranged from 19.0 to 21.5°C, dissolved oxygen ranged from 5.1 to 7.5 ppm, and salinity ranged from 

 24.0 to 28.5 ppt in the test samples and 29.5 to 32.5 ppt in the LIS controls. The 96-hr LC50 for the 

 reference toxicant, cadmium chloride, was 0.33 mg/L as chloride. Mean survival (n=5) in the LIS 

 control was 89±5.8% (range of 85-100%). 



The amphipod survival in the control sediments ranged from 79% to 95% in the six test series. Usually, 

 acceptable amphipod survival in controls is 85% or greater. However, the data from test series 5, in 

 which survival was 79%, were accepted since survival in all the test samples was either very high or 

 very low. The results would not have changed significantly if the samples had been retested. 



Amphipod survival ranged from 0.0% in two samples to 100% in one sample (Table 7). In 48 (84%) of 

 the 57 samples that were tested, mean amphipod survival was 80% of controls or less. In 46 (96%) of 

 the 48 samples in which amphipod survival was 80% of controls or less, the results were significantly 

 different from the controls. 



Table 7. Mean percent amphipod (A. abdita) survival in the 1993 Newark Bay survey performed 

 during Phase 2. 



Mean % 



survival 



± std. dev. 



95.0±5.0 

 95.015.0 



Percent 

 of Control 



100 

 100 



Significantly 

 less than control 

 (alpha=0.05) 



<80% of 

 Control 



36 



